CN109075010B

CN109075010B - Cold cathode ionization vacuum gauge and method for measuring pressure

Info

Publication number: CN109075010B
Application number: CN201780027137.8A
Authority: CN
Inventors: T·C·斯文尼; C·L·珀西; D·W·马里昂; B·J·凯利
Original assignee: MKS Instruments Inc
Current assignee: MKS Instruments Inc
Priority date: 2016-05-02
Filing date: 2017-04-27
Publication date: 2020-09-15
Anticipated expiration: 2037-04-27
Also published as: KR20190003669A; TWI725180B; JP2019515293A; JP6826131B2; KR102414085B1; SG11201809453SA; DK3443579T3; US10337940B2; EP3443579A1; WO2017192352A1; US20170315012A1; EP3443579B1; TW201741639A; CN109075010A

Abstract

A cold cathode ionization gauge and a method of measuring pressure. A cold cathode ionization gauge includes a plurality of cathodes with different spacing between the plurality of cathodes and an anode. The multiple cathodes allow pressure measurement over a wide pressure range. The first cathodes with larger spacing may provide current according to a townsend discharge; while a second cathode with a smaller spacing may provide current according to both a townsend discharge at higher pressures and a paschen law discharge at higher pressures. The feature on the second cathode may support paschen's law discharge. The large resistance between the cathode and the return of the power supply enables control of the output curve to expand the pressure range with accurate response and avoid output minima. Pressure measurements may be obtained from the output of the cathodes throughout a wide pressure measurement range based on the current from each cathode. The plurality of cathodes may also provide measurements that avoid discontinuities found in the current output of each cathode.

Description

Cold cathode ionization vacuum gauge and method for measuring pressure

RELATED APPLICATIONS

This application claims the benefit of U.S. provisional application No. 62/330,308 filed on 5/2/2016. The entire teachings of the above application are incorporated herein by reference.

Technical Field

The present invention relates to a cold cathode ionization gauge, and more particularly, to a cold cathode ionization gauge having a plurality of cathodes and a method of measuring pressure.

Background

Cold cathode ionization vacuum gauges (CCIG) are well known. Three conventional types of CCIGs include a common (non-inverting) magnetron vacuum gauge, an inverting magnetron vacuum gauge, and a Philips (or penning) vacuum gauge. All of these types of vacuum gauges have at least two electrodes (i.e., an anode and a cathode) within an evacuated non-magnetic enclosure that is connected to the vacuum to be measured. A high dc voltage potential difference (voltage potential) is applied between the anode electrode and the cathode electrode to generate an electric field between the electrodes. A magnetic field is applied along the axis of the electrode perpendicular to the electric field to lengthen the free electron path to maintain a pure electron plasma in which electrons collide with molecules and atoms to produce ions. The ions move to the cathode electrode to maintain the discharge current at a steady state value that is a function of pressure.

CCIG provides an indirect measure of the total pressure of a vacuum system by first ionizing the gas molecules and atoms within its gauge enclosure and then measuring the resulting ion current. The measured ion current is directly related to the gas density and total gas pressure inside the vacuum gauge enclosure, i.e. as the pressure inside the vacuum system decreases, the measured ion current also decreases. The gas specific calibration curve provides the ability to calculate total pressure from ion current measurements.

The CCIG described herein relies on the inverse magnetron principle. The vacuum gauge is cylindrically symmetric. The large voltage potential gradient (i.e., radial electric field) between the anode pin (on the shaft) and the cathode cylindrical enclosure energizes the electrons to cause an ionization event to occur. The crossed axial magnetic field provides the long electron trajectory path length required to sustain a pure electron plasma inside the enclosure. The discharge current is a measured quantity proportional to the pressure within the system.

The discharge is established by an avalanche ionization process, which typically begins with a single electron being released into the ionization volume of the vacuum gauge. The process responsible for releasing electrons may include a field emission event or a cosmic ray ionization process. The avalanche process (avalanche process) relies on a long path length for the electron trajectories, which results in many ionization processes per electron. Each ionization process releases ions and additional electrons that are added to the discharge. When ions collide with the inner wall of the cathode, additional electrons are also released into the discharge, thereby contributing to the total charge. As a result of the crossing electric and magnetic fields, a pure electron plasma is established around the anode as a sheath (sheath). The electron density and pressure are substantially independent. Ionization of neutral gas molecules occurs mainly inside the electron sheath with a constant pressure. All ions generated are guided by the electric field to the cathode and are hardly affected by the magnetic field. The ion current obtained is related only to the electron density inside the sensor and the total pressure of the gas.

The dual inverting magnetron design of Brucker et al, U.S. patent application No. 14/500,820, U.S. patent publication No. 2015/0091579, shown in fig. 1A, includes two

magnets

115a, 115b held together within a magnet assembly, with the poles of the two magnets facing each other. The dual inverting magnetron provides some of the characteristics of maximum magnetic field and thus maximum vacuum gauge sensitivity that can be achieved. A large gauge sensitivity is necessary to be able to operate at ultra-high vacuum (UHV) levels (i.e., pressures less than about 10^-9Torr (Torr) and as small as 10^-11Torr) to read a reliable pressure. U.S. patent application No. 14/500,820 is incorporated herein by reference in its entirety.

In a CCIG of the inverted magnetron type, a small leakage current may flow directly from the anode 110 to the cathode 120 via the inner surface of the gauge, and the presence of a so-called "guard ring" is known to collect this leakage current, thereby preventing it from reaching the cathode electrode and being detected by the gauge itself. To perform this function, the guard ring is electrically isolated from the cathode electrode and is typically held at a small positive voltage potential difference relative to the cathode electrode.

As shown in fig. 1A, 1B and 1C, the CCIG 100 includes a feedthrough 101, the feedthrough 101 including a grommet connection 102 that provides electrical connection to a grommet electrode 140 described below. Within the grommet connection 102, the anode grommet insulator 106 provides electrical insulation around the anode connection 110a of the elongated anode electrode 110. The grommet electrode 140 is connected to an initiator device 150, which will be described below. The grommet connection 102 is connected to a welding surface 104 through a cathode-grommet insulator 103, the welding surface 104 being seam welded to a unitary flange assembly 105, which will be described below. As shown in fig. 1A and 1B, the unitary flange assembly 105 includes an outer flange 105a and an inner flange 105B. The inner flange 105b surrounds the cathode electrode 120, which cathode electrode 120 surrounds the anode electrode 110 along its length and forms a discharge space 130 between the anode electrode 110 and the cathode electrode 120. The baffle (shown in fig. 1A as two

separators

170 and 180 having

holes

175 and 185, respectively) is connected to the cathode electrode 120.

As mentioned above, the crossed axial magnetic field provides the electron trajectory path length required to sustain a discharge inside the discharge space 130. The magnetic field is generated by magnet assembly 115, as shown in FIGS. 1A and 1B. The magnet assembly 115 includes a ferromagnetic spacer 114. The magnet assembly 115 may also include aluminum (or other non-magnetic material) spacers 113 at its ends closest to the shroud ring 140 to adjust the position of the discharge exit shroud ring 140.

A conductive grommet electrode 140 is interposed between the cathode electrode 120 and the anode electrode 110 around the base of the anode electrode 110 to collect leakage current that would otherwise tend to flow between the anode electrode 110 and the cathode electrode 120 if conductive deposits were to accumulate on the surface of the cathode-grommet insulator 103 exposed in the discharge space 130 for a long period of time during operation of the vacuum gauge 100.

A discharge initiator device 150 is disposed over and electrically connected to the guard ring electrode 140. As shown in fig. 1B, the starter device 150 has a plurality of tips 160 facing the anode 110 and forming a gap between the tips 160 and the anode 110 (3 tips are shown in a cross-sectional, cylindrically symmetric view in fig. 1B). The gap between the tip and the anode may be in the range of about 500 μm to about 2500 μm. The gap is configured such that when a voltage potential difference between the starter device 150 and the anode 110 is established, the field emission current is in the range of about 1 picoampere (pA) to about 10pA during normal operation. The amplitude of the field emission current depends on several parameters, such as the voltage potential difference, the size of the gap, the number of tips on the starter device and the kind of material from which the starter device is made. During operation of the CCIG, the voltage potential difference between the starter device and the anode may be in the range of about 0.4 kilovolts (kV) to about 6kV, for example, about 3.5 kV. This voltage potential difference generates electrons by field emission from the tip 160 to the anode, thereby seeding some electrons into the discharge volume 130 to trigger the avalanche process responsible for establishing the discharge. Alternatively, the voltage potential difference between the starter device and the anode may be configured to increase from about 3.5kV to about 5kV during start-up of the vacuum gauge to increase the field emission current by momentarily increasing the high voltage supply bias supplied to the anode electrode until a discharge is detected by a sudden increase in discharge current.

As shown in the electronic controller of fig. 1C, a limiting resistor 410 is provided between the anode electrode 110 and the high voltage power supply 430 (HVPS). The role of the limiting resistor 410 is to set an upper limit value of the amount of discharge current that can flow through the discharge volume 130 and to extend the life of the vacuum gauge. The result of the limiting resistor 410 is that the actual high voltage bias present at the anode electrode 110 and measured by the voltmeter 420 is generally less than the voltage provided by the HVPS 430. In fact, even though the output of the HVPS430 remains constant at all pressures, the anode voltage decreases as the ion current increases with pressure. In the vacuum gauge described herein, a 25 megaohm (M Ω) limiting resistor 410 was chosen to provide several benefits: 1. there is a safe limit to the amount of current that an HVPS can provide to an individual element in the event of accidental contact with an internal HVPS element, 2. the selection of resistors moves the pressure curve discontinuity above 1x10^-6The higher pressure range of torr, and 3. the upper limit of the discharge current is 125 mua when the anode voltage is set to 3.5 kV. The processor 490 at the CCIG controller continuously measures the anode voltage V with the voltmeter 420 and the discharge current I with the ammeter 460_DTo calculate the discharge impedance Z as a function of pressure, the processor ensures that the output of HVPS430 is constant over the entire pressure range. The processor also measures the guard ring current using current meter 470To monitor the amount of leakage. By this circuit configuration, the two independent current loops ensure that the leakage current at the anode feed-through does not cause any inaccuracies in the pressure measurement that depend on the discharge current impedance measurement.

CCIG is typically limited to less than 10^-2Operation in the low pressure range of torr. To measure pressures extending up to the range of atmospheric pressure (760 torr), the CCIG may be combined with a pressure gauge using a different technique, such as a thermal conductivity or diaphragm pressure gauge.

Disclosure of Invention

According to the invention, a CCIG is provided having a second cathode and the pressure is determined from the measured current flowing to each cathode. By means of different dimensions and electrical connections, different potential gradients (electrical potential gradients) are obtained between the common anode and the individual cathodes.

As in conventional CCIG, the magnet applies a magnetic field via at least a first spacing between the anode and the first cathode to lengthen the free electron path and thereby sustain the plasma between the anode and the first cathode and the generated ionic current flowing into the first cathode. This discharge is called Townsend discharge (Townsend discharge). The current measured from the cathode is typically allowed to be at low pressure (e.g., below 10 f)^-2Down to 10^-11Pressure in torr).

A second cathode electrically isolated from the first cathode and spaced from the anode by a second spacing less than the first spacing enables discharge and pressure measurement at a pressure higher than the low pressure required to form a plasma in the first spacing. The electrical controller applies a voltage between the anode and each of the first and second cathodes to produce plasma discharge ionization at a low pressure at least between the anode and the first cathode and discharge ionization at a pressure above the low pressure between the anode and the second cathode. The discharge between the anode and the second cathode may also include a thomson discharge similar to the first cathode, but at a higher pressure. Alternatively, it may be preferable additionally to allow Paschen's Law discharge when there is breakdown (breakthrough) between the anode and cathode. The controller determines the pressure based on the measured current flowing to the first cathode and based on the measured current flowing to the second cathode.

The controller may additionally measure the anode voltage, calculate the impedance between the anode and each cathode, and determine the pressure from these impedances.

In the disclosed embodiment, each cathode surrounds the anode and is cylindrical, with the different spacing being determined by the respective radius of the cylindrical cathode. For example, a taper may be provided at the opening of the second cathode to vary the electric field. In the disclosed embodiment, only two cylindrical cathodes are provided, but additional cathodes may be provided which provide additional spacing for additional measurements.

In a typical ionization gauge, the cathode provides a spacing of about 10 millimeters (mm) from the anode over a length of about 25mm, and similar dimensions are also suitable for the first cathode. Based on standard design, the spacing between the anode and cathode (i.e., the first spacing in the disclosed embodiment) is in the range of 5 to 15mm, and the first cathode has an effective length inside the magnet along the anode in the range of 15 to 40 mm. The smaller spacing between the anode and the second cathode should typically be in the range of 1.0 to 5.0mm, for example 2.4mm, to sustain the ion generating plasma at a higher pressure. The second cathode should have a length of at least about 6.0mm to sustain a townsend discharge. To prevent an oversized vacuum gauge, the length of the second cathode in the direction along the anode should be less than 24 mm. In the disclosed embodiment, the second cathode is about 16mm long.

The first cathode and the second cathode may be disposed within a polymeric housing that electrically isolates the cathodes from each other and from ground.

In order to measure even higher pressures than can be measured from a townsend discharge in a second, smaller cathode extending along the anode, features oriented towards the anode may be provided on the second cathode to establish a narrow gap between the anode and the cathode at the features. This feature allows for breakdown discharge at high pressures near atmospheric pressure according to paschen's law. A suitable gap between the anode and the feature is in the range of 0.3 to 1.0mm, with a preferred gap at an anode voltage of about 3kV of about 0.6 mm. The optimum gap is voltage dependent and the anode voltage of a standard CCIG ranges from about 2kV to 6 kV. At higher voltages, gaps at the higher end of the preferred range may be used. For cathodes operating with paschen's law discharge, the slope of the current versus pressure response is determined at any particular voltage gap. When the gap is large, the current response is very limited over the pressure range or the arc discharge will break down to block the pressure measurement; when the gap is small, the current and impedance response have a lower slope magnitude, sensitivity decreases and measurement error will increase.

The paschen law discharge feature configuration may be a disk with holes in the disk to form gaps. Alternatively, the feature may be a disk having one or more points (points) extending from the disk toward the anode. The feature may be one or more pins extending inwardly from the cathode cylinder. In one embodiment, the feature is a threaded pin that is inserted from the outside of the cathode and allows adjustment of the gap between the anode and the bottom end of the pin. The features may also be oriented on the anode toward the cathode.

A large resistance of about 500 kilo-ohms (k Ω) between each cathode and the return (return) connected to the power supply supplying voltage to the anode reduces fluctuations (random noise and oscillations) in the output response. Further noise reduction is achieved by an impedance of at least mega ohms (M Ω) between the cathode and the return of the power supply. To improve the slope of the current and impedance response, avoiding both steep and flat responses, the impedance connected to the second cathode is at least one order of magnitude greater than the impedance connected to the first cathode. In one embodiment, a resistor greater than 1.5M Ω is connected to the first cathode and a resistor greater than 30M Ω is connected to the second cathode. At least one of the impedances, in particular the second cathode impedance, may be provided by a variable resistor.

As mentioned earlier, the impedance measurement can be calculated using the anode voltage rather than just the cathode current. Different algorithms may be selected to provide the pressure output. The algorithm may be selected based on the impedance measurement and may have the impedance measurement as its input. The algorithm may be processed with pre-computed data stored in a look-up table.

The pressure may be determined from the electrical output from each of the first cathode and the second cathode over different pressure ranges, including non-adjacent pressure ranges over which the pressure is determined from the first cathode output. For example, the pressure may be determined based on the output of the first cathode at a first low pressure range, based on the output of the second cathode at a second pressure range higher than the first pressure range, based on the output of the first cathode at a third pressure range higher than the first and second pressure ranges, and based on the output of the second cathode at a fourth pressure range higher than the first, second, and third pressure ranges.

In preferred operation over an extended pressure range, at least a townsend plasma discharge is supported between the anode and the first cathode at low pressure; at least a Thomson discharge or a Paschen's law breakdown discharge is supported between the anode and the second cathode at a pressure above the low pressure. A breakdown discharge may be supported at the feature of the second cathode.

In a method of measuring pressure, a magnetic field is applied to a first space between an anode and a first cathode. At low pressure, electrons are released into the first space to generate a Thomson plasma discharge and an ion flow to the first cathode in the first space. At a pressure above the low pressure, a discharge is generated between the second cathode and the anode to generate a current flowing to the second cathode. The pressure is based on the measured current flowing to the first cathode and the measured current flowing to the second cathode. The discharge between the second cathode and the anode may be a thomson plasma discharge or a paschen law breakdown discharge, or it may include each of these discharges at different pressure ranges.

In another method of measuring pressure, electrons are released into a first space to create a plasma discharge within the first space and a flow of ions to a first cathode, the current response of the first cathode with respect to pressure having a first discontinuity. Electrons are released into the second space to create a plasma discharge within the second space and a flow of ions to the second cathode, the current response of the second cathode with respect to pressure having a second discontinuity. The pressure is determined based on the measured current flowing to the first cathode and the measured current flowing to the second cathode. The pressure is based on the current flowing to the first cathode measured through the pressure including the second discontinuity and the current flowing to the second cathode measured through the pressure including the first discontinuity.

Drawings

FIG. 1A illustrates a prior art CCIG to which the present invention may be applied.

FIG. 1B is an enlarged perspective view of the anode and starter ring of the vacuum gauge of FIG. 1A.

FIG. 1C illustrates an electrical controller associated with the vacuum gauge of FIG. 1A.

Fig. 2 illustrates the current output from the cathode and the guard ring when the vacuum gauge of fig. 1A-C is modified in accordance with the present invention.

FIG. 3 illustrates a modified control of the vacuum gauge of FIG. 1A.

Figure 4 is a longitudinal cross-section of a modified CCIG embodying the present invention.

FIG. 5 illustrates the current and voltage output of the vacuum gauge of FIG. 4.

Fig. 6A-6E illustrate alternative configurations of cathode features for inducing paschen's law discharge.

Fig. 7 illustrates a preferred embodiment of the present invention.

FIG. 8 illustrates an electrical controller for use with several embodiments of the present invention.

Fig. 9A and 9B illustrate impedance measurements across a wide range of pressure measurements with different cathode resistors, depending on cathode current and anode voltage, and fig. 9B illustrates a graph used in an algorithm to determine pressure.

Fig. 10A and 10B illustrate a process used by processor 802 to determine pressure from the impedance measurements of fig. 9B.

FIG. 11 illustrates an alternative use of the present invention to avoid breaks in the vacuum gauge output.

Fig. 12 illustrates an alternative embodiment of the invention specifically designed to avoid breaks.

FIG. 13A is a graph of impedance measurements similar to FIG. 9B but for a device having different output characteristics; and FIG. 13B is a flow chart similar to FIG. 10B but for an apparatus having the characteristics of FIG. 13A.

Detailed Description

The following is a description of exemplary embodiments of the invention. The teachings of all patents, published applications and references cited herein are incorporated by reference in their entirety.

It has been determined that the vacuum gauge shown in U.S. patent application No. 2015/0091579 can be modified with an electronic controller that operates the vacuum gauge at higher pressures up to 1 torr or even atmospheric pressure of 760 torr by maintaining a high voltage at the anode at the higher pressure described below and modifying the electronics of the detector (particularly processor 490) to rely not only on the cathode current but also on the current flowing through the guard ring.

Fig. 2 shows the cathode current in this design, as well as the current of the guard ring, which now serves as the second cathode. In the general operation of the vacuum gauge, the pressure is determined from curve 200 and is about 10 above 201^-2The anode voltage is turned off at torr pressure. Above this pressure, the current output 202 levels out so as to become less reliable and drops rapidly at 204 of about 1 torr. The guard ring current is very low and is only used to monitor leakage currents. However, if the operation of the vacuum gauge is maintained at a higher pressure as shown in FIG. 2, the grommet current can be seen to rise at 206 as the cathode current drops. Thus, at higher pressures, the grommet current may be used as an indication of pressure. It can be seen that at about 10 torr, the grommet current begins to drop at 208. See only over 1 torr to 1x10³Guard ring currents of the torr will give ambiguous results. For example, a current of 80 microamperes (μ A) would represent 8 Torr or more than 100 Torr. However, by also monitoring the cathodic current, one can monitor the cathodic current even at those high pressuresIt will be known on which side of the wave crest the guard ring current flows.

Since the vacuum gauge of the cited patent application is not designed for high pressure operation, it is between 10^-2The flat region to 1 torr still lacks accuracy and there is difficulty monitoring the grommet current at the peak grommet current, but it does provide the opportunity to measure pressure over a much wider pressure range without the need for additional pressure gauges (e.g., resistance or diaphragm gauges). For increased current in the guard ring, the principal cathodic Thomson discharge is likely to migrate (transition) to the region of the guard ring below the starter. Paschen's law breakdown is likely to occur at the starter tip 160 after the grommet current peak, where the gap to the anode is much shorter (i.e., 0.676 mm).

The operation of the vacuum gauge described above over a wide pressure range can be seen in fig. 3. Unlike prior art vacuum gauges, at 302, the high voltage penetration includes greater than 10^-2A wide range of pressures above 1 torr, and preferably up to atmospheric pressure, is applied to the anode. Current from the cathode is detected at 304 and current from the grommet is detected at 306. Optionally, the anode potential (anode potential) may also be detected at 308. At 310, the current from the cathode, the current from the grommet, and possibly the anode potential are processed to determine the pressure. Previously, the current from the guard ring was used only to monitor the leakage current. As described above, this process is not simply a conversion from current (e.g., cathodic current) to pressure. Instead, multiple currents must be monitored and a decision must be made as to which current to use to provide pressure based on the currents. The rising grommet current may indicate that the cathode current is to the right of the toggle point at 204 to distinguish the high pressure measurement from the low pressure measurement. Alternatively, the grommet current itself may be used to provide a pressure measurement above the toggle point 206. Although only current is shown, anode voltage may be used, or anode voltage and current may be used to calculate impedance, and impedance may be used to represent pressure.

FIG. 4 illustrates a modified vacuum gauge configuration to support the implementation of CCIG over a wide pressure range. Like the vacuum gauge disclosed in U.S. patent application No. 2015/0091579, the CCIG includes an anode 402 that extends axially into a cylindrical cathode (also referred to as a cathode cage) 404. Also, the cathode is surrounded by a magnet assembly 405 to sustain a free electron plasma. This differs from prior art vacuum gauge designs in that it includes an additional cathode cylinder 406 in series with the cathode 404 and surrounding the anode 402 remote from the gauge mouth (mouth) and flange 420. However, the diameter of the cathode 406 is smaller to provide a reduced spacing between the cathode and anode and the resulting different electric field gradient. The two cathodes are separated by an insulator 408. In this embodiment, both cathodes are surrounded by a magnet assembly 405. The vacuum gauge is optionally modified using a polymeric housing 411 in accordance with the method disclosed in PCT patent application No. US 2016/013219 filed 2016, 1, 13, 2016 (the entire contents of which are incorporated herein by reference). The cathode structure may be disposed within the housing 411 during a molding operation of the housing 411, wherein the

electrodes

402 and 414 extend through the polymer housing to the respective large cathode 404 and small cathode 406.

It has been found that the thomson plasma is located within the large cathode 404 at low pressures, but moves into the smaller cathode 406 at higher pressures. A taper 416 at the mouth of the small cathode 406 can support plasma transfer (transition) from the large cathode to the small cathode.

FIG. 5 illustrates at 10^-6Bracket to 10³A large cathode current 502, a small cathode current 504, a combined current 506, and an anode voltage 508 over a wide range of pressures. The inclusion of two chambers in a cold anode ionization gauge design allows for the delivery of reasonable discharge currents over a wide pressure range. As the measured pressure increases, the molecular density also increases, which causes the mean free path of the molecule-to-molecule collisions to shorten significantly. The use of two chambers with different geometries enables each chamber to be optimized to deliver a detectable discharge current for a desired molecular density and mean free path. The mechanism for delivering current through the gaseous medium is to generate a plasma, anAnd it has been observed that there is a specific pressure (in the range of 10-200 torr) that causes the plasma to become unstable (oscillate and randomly fluctuate), which causes the apparent impedance to vary greatly, which causes the discharge current to vary greatly. This makes it difficult to use the instantaneous discharge current measurement to determine the measured pressure. Because the voltage supply used on the CCIG is equipped with a current limiting resistor, it has been found that measuring the voltage after this current limiting resistor provides a more stable means of determining the discharge current within the device. This does not require sensitive measurement circuitry to accurately resolve the delivered current and does not require downstream processing or filtering to compensate for instabilities in the plasma. Because the measured voltage is relatively high (ranging from hundreds to thousands of volts), the design of this measurement circuit need not have the accuracy required to measure low currents (which are typically currents in the microamp range).

The operation of the vacuum gauge will be as shown in fig. 3, except that a second cathode 406 designed for this function replaces the guard ring in step 306.

The CCIG of FIG. 4 includes multiple cathodes; both cathodes have a common anode to which a high voltage is supplied. The separate currents detected from each cathode over the pressure range allows us to use these currents as a means of determining when to switch from one measurement scheme to another. Both measurement schemes use either current or resistance to derive pressure, or anode voltage alone to derive pressure. The separate currents from each cathode enable us to detect whether it is in a low pressure region (<0.1 torr) or a high pressure region (>0.1 torr).

When the measured pressure is rising or falling, the current delivered to the two cathodes will be measured and an algorithm will be used to determine whether the value measured from the first or second cathode (which may be a voltage or current) will be used as a basis for calculating the measured pressure. In one design, at low pressures (below 0.1 torr), the total current or sum of currents (because there is current from both cathodes) can be used as the basis for calculating the measured pressure, and at pressures of about 0.1 torr and higher, the measured current is switched to only the smaller cathode, and we change to measuring the anode voltage because the response of the current to pressure has a small slope and has low sensitivity. Thus, by detecting the magnitude of the current in each cathode, we can determine which pressure range is within, and thus can determine whether the current or voltage should be used to calculate the pressure.

There are several important points to consider:

1. the individual cathode shield currents are to be considered and fig. 5 shows that small and large shield currents have their own characteristics associated with pressure.

2. There is also an anode voltage that can be used for the measurement.

3. The cathodic current may be used alone to determine the pressure range and/or may also be added to improve the linearity and/or sensitivity of the measurement.

4. The current and voltage may be combined to provide a discharge impedance.

FIG. 4 also illustrates another modification of the vacuum gauge structure to support Paschen's law discharge at higher pressures near atmospheric pressure. Feature 418, in this example a cathode pin, possibly in the form of a set screw, within the inner surface of small cathode 406, provides a small spacing between cathode 406 and anode 402. The preferred spacing is in the range of 0.3 to 1.0 mm.

The upper pressure range measurements (1-760 torr) for CCIG present technical challenges not present at pressures below 1 torr: a mode change associated with pressure, an oscillating discharge behavior, and a non-Thonsen discharge characteristic. One factor that causes significant instability is the shift in the location of partial discharges across the high voltage cathode shield. This threaded cathode pin feature configuration provides a way to control the plasma position in the region of paschen's law arc discharge, eliminating spatial positional instability and current and voltage spikes (spikes) that would otherwise be caused if the discharge were allowed to move around the cathode. This also gives us a way to set the operating electric field between the anode and cathode pins by adjusting the gap and voltage between the anode and cathode to ensure that there is a sufficient potential gradient to operate at all times up to or even greater than 1 atmosphere.

At pressures between about 1 torr and 760 torr, the CCIG plasma discharge tends to change from a uniform glow (glow) to a small confined electric light (bolt) that jumps around within the interior space of the cathode housing (arc discharge region). Without the feature configuration described in this disclosure, the electro-optic discharge continuously changes position around the cathode housing. The purpose of the cathode pins is to control the spatial position of the discharge and to minimize spatial fluctuations in the process.

The spatial fluctuations of the electro-light cause the result of anode voltage and cathode current spikes and/or oscillations. Minimizing the effects of large spatial fluctuations of the discharge smoothes the current and voltage to yield a more reproducible device with a simple signal output for analysis.

The cathode pin-anode spacing is used to set and establish the breakdown voltage at higher pressures. The breakdown voltage is described as the minimum voltage at which the discharge current is detected. For nitrogen, there is a minimum on the plot of breakdown voltage versus pressure between 0.1 torr and 10 torr (plot). In order for a vacuum gauge to operate properly at high pressures in the region of paschen's law, we must always operate at a voltage above the breakdown voltage to obtain a signal from which the pressure can be derived. The cathode pin-anode spacing is set such that discharge naturally occurs first between the cathode pin and the anode, since the spacing is the shortest path between the two.

The threaded cathode pin allows small adjustments to be made to the distance between the anode and cathode. When an appropriately designed and arranged feature configuration is used, the arc position is fixed and the discharge oscillation amount is reduced. The arc is visible to the naked eye.

In the high-voltage region, the local arcing can cause severe wear on both the anode surface and especially the cathode surface due to sputtering. This is why alternative materials of construction are considered. Stabilizing the arc position helps to minimize signal fluctuations and concentrate sputtering damage at a single point. The choice of the structural material for the cathode pin is worth careful consideration as it has a direct impact on the long-term stability of the reading. Sputtering from the cathode features may be minimized by using known sputter hardening materials (e.g., iridium, titanium, tungsten, or alloys of these materials) for the cathode features. Adding geometric features to the feature configuration also compensates for wear and extends life.

Sputtering damage can also be mitigated at high pressure by adjusting the anode voltage to reduce the duty cycle from continuous (100%) to a smaller ratio (< 100%). This allows the anode material to cool between discharge events and prevents the anode surface from melting.

At higher pressures, the arc discharge selectively falls on the shortest path provided by the cathode pin, since that path has the greatest potential gradient. At lower pressures dominated by the Thomson discharge (spatially uniform discharge), the cathode pins will have little or no perturbation to the discharge (perturbation). By controlling the arc, we can limit or reduce the current and voltage spikes we see in this pressure region; controlling the minimum distance controls the breakdown voltage and the location of the discharge. Controlling the location where the arc occurs and the voltage at which it occurs can stabilize operation and minimize sputtering. Moreover, it allows for a larger diameter high pressure cathode housing to improve sensitivity in the intermediate pressure range. At higher pressures, the discharge is driven strictly by static electricity, and magnetic fields have little effect on their behavior.

The threaded cathode pin 418 can be seen in the end view of fig. 6E. Other cathode feature configurations used to maintain the proper gap between the small cathode 406 and the anode include holes 602 in the disk 604 shown in fig. 6A, tips 606 in the disk 608 of fig. 6B, and pins 612 in fig. 6C. Fig. 6D illustrates a protrusion 610 protruding from the wall of the small diameter cathode 406.

A preferred embodiment of the vacuum gauge is shown in fig. 7. Here, a large diameter cathode 702 and a small diameter cathode 704 are molded around an anode 707 into a polymer housing 706 that provides isolation between the two cathodes. Electrode 708 extends to small diameter cathode 704 and electrode 709 extends to large diameter cathode 702. Alternatively, either electrode may extend through one side of housing 706. Because the magnetic field is less important in supporting a townsend discharge in a small diameter cathode, magnet 714 only surrounds a large diameter cathode. A flange 710 is formed at an open end of the vacuum gauge to be coupled to a chamber whose pressure is to be monitored.

Specific dimensions and preferred dimensional ranges are also provided in fig. 7. For each dimension, the actual dimension in millimeters is provided within brackets. Actual dimensions in inches are provided below it and preferred ranges in millimeters are provided below it. The length of the large cathode is the effective length, i.e., the length that is located within the magnet 714 and exposed to the gas environment. Although the cathode cylinder 702 extends beyond the magnet, the plasma with ionization phenomena is primarily localized in the region inside the magnet. The effective length of the small cathode that is not within the magnet is its entire length. For the high voltage region (i.e. the region within which the cathode operates), extending the magnet beyond the cathode will have little effect. The magnetic field from magnet 714 will extend into the space, but the additional magnetic field from the additional magnet surrounding the cathode will not be effective.

The paschen law feature configuration 716 is of the type shown in fig. 6A and provides a clearance in the range of 0.3 to 1.0mm, and in particular, 0.6mm (0.024 inches). Any of the other feature configurations of fig. 6B-6E may also be used. As previously described, this feature 716 supports arcing at pressures near atmospheric pressure. Although the feature may be disposed toward the proximal open end of second cathode 704, the feature is disposed toward the distal end of second cathode 704 in order to leave the mouth of the cathode open for plasma discharge. For molding purposes, the feature is shown distally.

The magnets used in this embodiment are approximately 800 or 900 gauss, which is in the preferred range of 500-1100 gauss.

FIG. 8 illustrates controller electronics for the vacuum gauge in any of the embodiments. As in the prior art circuit, the power supplied to the anode is supplied from the power source P to the anode via a current limiting resistor R1, which may be a high resistance (e.g., 30M Ω) resistor. The voltage supplied to the anode is sensed by a voltage sensor V1 connected to a voltage divider R2, R3 and to ground. Alternatively, the anode voltage may be determined from the voltage output of the power supply P and the measured voltage drop across the resistor R1. The current from the large cathode 702 and the electrode 709 is detected by means of a voltage sensor V3 connected across a current sense resistor R5 (which may be, for example, about 50k Ω). The current from the small cathode and its electrode 708 can be detected with a voltage sensor V2 connected across a current sense resistor R4 (which can, for example, also be about 50k Ω). The processor 802 controls the power supply, varies any variable resistance, and receives the sensed signal. It also outputs a pressure reading, which will be described with reference to fig. 10.

Of particular importance in this circuit is an additional resistor RS connected to the small cathode 704 and a resistor RL connected to the large cathode 702. In the original design, both RL and RS used 523k Ω resistors to cancel noise in the output signal from the cathode and filter oscillations. However, with much larger resistances in excess of 1M Ω, the shape of the current output relative to pressure can be controlled to provide more accurate pressure readings. In detail, the resistor RS connected to the small cathode is raised to 30M Ω, and the resistance RL is raised to 1.27M Ω or 2.07M Ω. With the extremely high resistor at the small cathode, more current is pushed into the large cathode to increase the slope magnitude at high pressures above 1 torr. To allow the pressure response to be dynamically controlled with changing conditions (e.g., different gas species), resistors RL and RS, and in particular RS, may be variable resistors.

Fig. 9A and 9B illustrate the output impedance calculated as the ratio of anode voltage to cathode current for each cathode of the device of fig. 7 using the circuit of fig. 8. FIG. 9A is 523k Ω for RS and RL, respectively. FIG. 9B is for RL being equal to 1.27M Ω and RS being equal to 30M Ω. The adjustment of the cathode resistance also enables shifting (shifting) the point of local minima in the respective currents and impedances to produce a sensor in which there are no regions within the sensed signal where pressure cannot be resolved and displayed with sufficient accuracy.

Five different operating regions can be identified in fig. 9B. Below 10^-2At low torr, the vacuum gauge operates much like a standard CCIG with a large cathode Thomson plasma discharge. Between 10 at still low pressure^-2To the region of about 1 torr, resistors RS and RL are used to extend the operation of the thomson plasma discharge of the large cathode. In the region of about 1 torr to about 35 torr, the plasma activity in the small cathode increases (impedance drops) and the plasma shifts (shifting) from the large cathode to the small cathode. Here, as the impedance of the small cathode decreases, the impedance measurement of the large cathode begins to increase. As the transition continues, from about 35 torr to about 89 torr, the plasma of the small cathode begins to turn off and the current reaches that caused by the breakdown voltage because the current of the large cathode continues to decrease causing high impedance. Finally, at pressures above about 89 torr to about atmospheric pressure (760 torr), the current discharge is a paschen's law arc discharge across the small gap provided by the additional features provided on the small cathode.

The processing of the algorithm of fig. 9B may be as shown in fig. 3, but the current detected in step 306 is the current of the second cathode designed for this function. Fig. 10A illustrates a more detailed process. At 1020, an anode voltage is applied to the anode throughout a large pressure range. At 1022, current from the first cathode is detected and at 1024, current from the second cathode is detected. At 1026, the anode voltage is detected. A more detailed description of step 310 for processing the current and potential to determine pressure is given in 1028, 1030, and 1032. At 1028, the cathode current and anode potential are processed to determine the impedance of each cathode. At 1030, one of these impedances is selected for further processing based on the pressure region. As shown in fig. 9B, the impedance shown in solid lines is chosen for each of the four pressure zones. At 1032, the selected impedance is converted to a pressure. In one embodiment, the selecting and converting

steps

1030 and 1032 are performed as shown in FIG. 10B.

Fig. 10B illustrates controller processor logic for accessing each of four different look-up tables to convert the impedance of the large cathode or the impedance of the small cathode to an output pressure. The four lookup tables correspond to the four solid lines of the impedance diagram of fig. 9B and are selected based on the impedances of the large and small cathodes. As shown in fig. 9B, the pressure is determined from the impedance of the large cathode at a low pressure corresponding to the standard CCIG and the extended standard CCIG. In this pressure region, decision block 1002 directs the processor to lookup table 1 of 1010. Then, in the region of small cathode plasma activity of about 0.5 to 35 torr, the system switches to follow the impedance of the small cathode to determine the pressure. Decision block 1004 directs the processor to table 2 at 1012. Then, to avoid seeing a minimum in the impedance of the small cathode at about 35 to 89 torr, the system switches back at 1006 to determine the pressure at 1014, following the impedance of the large cathode in look-up table 3. Finally, as the impedance of the large cathode approaches its peak, the system switches back to follow the impedance of the small cathode in table 4 at 1016. Thus, as can be seen in FIG. 10B, with each data sample, the system traverses decision blocks 1002, 1004, 1006, and 1008 to determine a lookup table to be used in 1010, 1012, 1014, or 1016 to identify the pressure for that data sample. After each lookup or if the decision tree cannot identify the lookup table, the system moves to the next sample.

Although the process has been described by way of example as moving from low pressure to atmospheric pressure, it will be understood that any data sample may direct the processor to any look-up table without regard to any pressure history.

FIG. 11 illustrates the wide pressure range design or more conventional pressure ranges (e.g., below 10) as described immediately above^-2Torr) to another use of multiple cathodes. CCIG typically has a discontinuity in its output. For example, a vacuum gauge having a single cathode may have an impedance that follows curve 1102, which is shown at 1104 with a discontinuity. At this break, it is difficult to provide an accurate pressure reading. By using two cathodes (the second having an impedance response following path 1106 (e.g. with a break at 1108)),breaks can be avoided when determining the pressure. At low pressures, the pressure may be determined by the impedance provided by the large cathode along curve 1110. The large cathode will be used for a pressure higher than the pressure at which the break 1108 occurs. However, at some pressure below the pressure at which the break 1104 would occur for a large cathode, the system would transition at 1114 to an impedance output that depends on the small cathode along curve 1112. Thus, interruptions are avoided.

The double cathode embodiment illustrated previously may be used to avoid discontinuities. Another embodiment is shown in fig. 12. Two electrically

isolated cathode sleeves

1202 and 1204 may be mounted within a housing 1205 of polymeric material (e.g., as previously described). The cathode surrounds the center of the anode 1206. Each cathode has a

respective ring magnet

1208, 1210 surrounding it. The magnet may be in a repulsive or "bucking" state such that two separate discharge regions are created, one for each cathode. The discontinuity can be moved to a different pressure range by changing the resistance at the cathode, changing the physical size of one cathode relative to the other, and changing the magnetic field. By knowing the response of each cathode, discontinuities can be avoided by having the transition 1114 of FIG. 11 occur at a predetermined pressure. Alternatively, by looking at the ratio of the currents of the two separate cathodes, it is possible to determine when a disruption will occur. In this case, another cathode current is used to inform the pressure. This is primarily due to the fact that there must be a reference pressure, and in turn one will be the reference pressure.

As can be seen from the above examples, the magnetic field can be established in many different ways. For example, fig. 1A illustrates a dual magnet that only surrounds a larger cathode. Fig. 4 illustrates a magnet assembly surrounding two cathodes. Fig. 7 illustrates a single magnet that only surrounds a large cathode. Fig. 12 illustrates separate magnets surrounding separate cathodes. This arrangement can also be used to extend the pressure range. It will be appreciated that other arrangements are possible.

FIG. 13A provides a diagram similar to FIG. 9B, but for devices with different output characteristics. The impedance of the small cathode is shown in dashed lines and the impedance of the large cathode is shown in solid lines. Fig. 13B illustrates a flow chart similar to fig. 10B, but for a device having the characteristics of fig. 13A. As previously described, at the lowest pressure, a look-up table for the large cathode is used to provide a pressure reading. At decision block 1302, if the impedance of the large cathode is determined to be greater than the impedance of the small cathode when the impedance of the small cathode is greater than 8E8, or if the impedance of the large cathode is less than the small impedance but the small impedance remains above 2.0E8, then a lookup table with the impedance of the large cathode is used to determine the pressure. Therefore, a large cathode look-up table is used for all pressures below 1318 in fig. 13A. If the condition of decision block 1302 is not met, but the impedance of the small cathode is found to be greater than the impedance of the large cathode at decision block 1306, then a look-up table for the small cathode is used to provide a pressure reading at 1308. Thus, the output of the small cathode is used between the pressures at 1318 and 1320 of fig. 13A.

If the conditions of decision blocks 1302 and 1306 are not met, decision block 1310 will determine whether the impedance of the large cathode is less than 3.0E 9. If so, then a look-up table of large cathodes is used at 1312 for the pressures in FIG. 13A between 1320 and 1322. Finally, if the conditions of decision blocks 1302, 1306, and 1310 are not met, then a look-up table for the small cathode is used for pressures above 1322 at 1314.

Once the pressure is determined by one of the look-up tables, the next data sample is collected at 1316 for evaluation as in fig. 13B.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the scope of the invention encompassed by the appended claims.

Claims

1. A cold cathode ionization gauge, comprising:

an anode;

a first cathode spaced from the anode by a first spacing sufficient to enable formation of a plasma between the anode and the first cathode at a first pressure and a resulting ionic current flowing into the first cathode;

a second cathode electrically isolated from the first cathode and spaced from the anode by a second spacing less than the first spacing, the second spacing enabling a discharge to be generated at a pressure higher than the first pressure required to form a plasma within the first spacing;

a magnet applying a magnetic field across at least the first gap to lengthen a free electron path to sustain the plasma;

an electrical controller applying a voltage between each of the first and second cathodes and the anode to produce plasma discharge ionization at the first pressure at least between the anode and the first cathode and discharge ionization at a pressure higher than the first pressure between the anode and the second cathode, the electrical controller determining the pressure from the measured current flowing to the first cathode and from the measured current flowing to the second cathode.

2. The cold cathode ionization gauge of claim 1 wherein the electrical controller determines a first cathode impedance and a second cathode impedance from the measured current flowing to the first cathode, the measured current flowing to the second cathode, and the measured anode voltage, and pressure is determined from the first cathode impedance and the second cathode impedance.

3. The cold cathode ionization gauge of claim 2 wherein a pressure is determined from the first cathode impedance and the second cathode impedance over each pressure range, the pressure ranges including non-adjacent pressure ranges in which the pressure is determined from the first cathode impedance.

4. The cold cathode ionization gauge of claim 1, wherein the first cathode and the second cathode surround the anode.

5. The cold cathode ionization gauge of claim 4 wherein each cathode is cylindrical and the different spacings are defined by respective radii.

6. The cold cathode ionization gauge of claim 5 comprising only two cylindrical cathodes.

7. The cold cathode ionization gauge of claim 5, wherein the spacing between the anode and the first cathode is in the range of 5 to 15 millimeters, and the first cathode has an effective length along the magnet and along the anode in the range of 15 to 40 millimeters.

8. The cold cathode ionization gauge of claim 7, wherein the spacing between the anode and the second cathode is in the range of 1 to 5 millimeters, and the second cathode has a length along the anode in the range of 6 to 24 millimeters.

9. The cold cathode ionization vacuum gauge of claim 8, wherein the second cathode includes features oriented toward the anode, the features establishing a gap between the anode and the features in the range of 0.3 to 1.0 millimeters to enable paschen's law discharge to occur between the anode and the features on the second cathode at high pressures near atmospheric pressure.

10. The cold cathode ionization vacuum gauge of claim 9, wherein the electrical controller comprises an impedance between the first cathode and a return to power of at least megaohms, and an impedance between the second cathode and the return to power that is at least one order of magnitude greater than the impedance between the first cathode and the return to power.

11. The cold cathode ionization gauge of claim 5, wherein the spacing between the anode and the second cathode is in the range of 1 to 5 millimeters, and the second cathode has a length along the anode in the range of 6 to 24 millimeters.

12. The cold cathode ionization vacuum gauge of claim 11, wherein the second cathode includes features oriented toward the anode, the features establishing a gap between the anode and the features in the range of 0.3 to 1.0 millimeters to enable paschen's law discharge to occur between the anode and the features on the second cathode at high pressures near atmospheric pressure.

13. The cold cathode ionization gauge of claim 5, wherein the first cathode and the second cathode are disposed within a polymer housing that electrically insulates the first cathode and the second cathode.

14. The cold cathode ionization gauge of claim 1 wherein the spacing between the anode and the second cathode is in the range of 1 to 5 millimeters and the second cathode has a length along the anode in the range of 6 to 24 millimeters.

15. The cold cathode ionization vacuum gauge of claim 14, wherein the second cathode includes features oriented toward the anode, the features establishing a gap between the anode and the features to enable paschen's law discharge to occur between the anode and the features.

16. The cold cathode ionization gauge of claim 15, wherein the gap between the anode and the feature is in the range of 0.3 to 1.0 millimeters.

17. The cold cathode ionization gauge of claim 15, wherein the feature is a disk and the gap is formed in a hole in the disk.

18. The cold cathode ionization gauge of claim 15, wherein the feature is a disk and the gap is formed between the anode and a tip extending from the disk.

19. The cold cathode ionization gauge of claim 15, wherein the feature is a pin.

20. The cold cathode ionization gauge of claim 15, wherein the feature is a threaded pin.

21. The cold cathode ionization vacuum gauge of claim 15, wherein the electrical controller comprises an impedance between the first cathode and a return to power of at least megaohms, and an impedance between the second cathode and the return to power that is at least one order of magnitude greater than the impedance between the first cathode and the return to power.

22. The cold cathode ionization gauge of claim 1 wherein the electrical controller comprises an impedance of at least megaohms between each cathode and the return to the power supply.

23. The cold cathode ionization gauge of claim 22, wherein the impedance from the second cathode is at least one order of magnitude greater than the impedance from the first cathode.

24. The cold cathode ionization gauge of claim 22, wherein at least one of the impedances is provided by a variable resistance.

25. The cold cathode ionization gauge of claim 22 wherein the electrical controller selects one of a plurality of algorithms to provide a pressure output based on the electrical measurements, the electrical controller selecting an algorithm based on the impedance measurements between the anode and each cathode.

26. The cold cathode ionization gauge of claim 25, wherein the algorithm is processed using pre-calculated data stored in a look-up table.

27. The cold cathode ionization gauge of claim 1, wherein pressure is determined from electrical output from each of the first and second cathodes over different pressure ranges, the different pressure ranges including non-adjacent pressure ranges in which the pressure is determined from the first cathode output.

28. The cold cathode ionization gauge of claim 27, wherein the pressure is based on an output of the first cathode in a first pressure range, based on an output of the second cathode in a second pressure range higher than the first pressure range, based on an output of the first cathode in a third pressure range higher than the first pressure range and the second pressure range, and based on an output of the second cathode in a fourth pressure range higher than the first pressure range, the second pressure range, and the third pressure range.

29. The cold cathode ionization gauge of claim 1 wherein at least a Thomson plasma discharge is supported between the anode and the first cathode at the first pressure and at least a breakdown discharge is supported between the anode and the second cathode at a pressure higher than the first pressure.

30. The cold cathode ionization gauge of claim 29 wherein the thomson plasma discharge is also supported between the anode and the second cathode at a pressure higher than the first pressure.

31. The cold cathode ionization gauge of claim 29 wherein the breakdown discharge is supported at a feature of the second cathode.

32. The cold cathode ionization gauge of claim 29 wherein the breakdown discharge is supported at a feature of one of the second cathode and the anode that reduces a spacing between the second cathode and the anode.

33. The cold cathode ionization gauge of claim 1, wherein at least a Thomson plasma discharge is supported between the anode and the first cathode at the first pressure, and at least a Thomson plasma discharge is supported between the anode and the second cathode at a pressure higher than the first pressure.

34. The cold cathode ionization gauge of claim 1 wherein each cathode is cylindrical surrounding the anode, the different spacing being determined by a respective radius, the diameter of the inner surface of at least one of the cathodes being tapered.

35. The cold cathode ionization vacuum gauge of claim 1, wherein the second cathode includes features oriented toward the anode, the features establishing a gap between the anode and the features to enable paschen's law discharge to occur between the anode and the features.

36. The cold cathode ionization gauge of claim 35, wherein the feature is displaced from the end of the second cathode near the first cathode.

37. A method of measuring pressure, the method comprising:

applying a magnetic field to a first space between an anode and a first cathode;

releasing electrons into the first space at a first pressure to create a plasma discharge within the first space and a flow of ions to the first cathode;

generating a discharge between a second cathode and the anode at a pressure above the first pressure to generate a current flowing to the second cathode; and

determining a pressure from the measured current flowing to the first cathode and from the measured current flowing to the second cathode.

38. A cold cathode ionization gauge, comprising:

an anode;

a first cathode spaced from the anode by a first spacing sufficient to enable formation of a plasma between the anode and the first cathode and formation of a resulting ionic current flowing into the first cathode, the current response of the first cathode with respect to pressure having a first discontinuity;

a second cathode electrically isolated from the first cathode and spaced from the anode by a second spacing less than the first spacing, the second spacing sufficient to enable formation of a plasma between the anode and the second cathode and a resulting ionic current flowing into the second cathode, a current response of the second cathode with respect to pressure having a second discontinuity;

a magnet applying a magnetic field across the first and second spacings to lengthen a free electron path to sustain the plasma;

an electrical controller applying a voltage between each of the first and second cathodes and the anode to produce plasma discharge ionization between each of the first and second cathodes and the anode, the electrical controller determining a pressure from a current flowing to the first cathode measured through the pressure including the second discontinuity and determining a pressure from a current flowing to the second cathode measured through the pressure including the first discontinuity.

39. A method of measuring pressure, the method comprising:

applying a magnetic field to a second space between the anode and a second cathode;

releasing electrons into the first space to create a plasma discharge within the first space and a flow of ions to the first cathode, the current response of the first cathode with respect to pressure having a first discontinuity;

releasing electrons into the second space to create a plasma discharge within the second space and a flow of ions to the second cathode, the current response of the second cathode with respect to pressure having a second discontinuity;

determining a pressure based on the measured current flowing to the first cathode throughout the pressure including the second discontinuity and based on the measured current flowing to the second cathode throughout the pressure including the first discontinuity.

40. A cold cathode ionization gauge, comprising:

an anode;

a first cathode spaced from the anode by a first spacing sufficient to enable formation of a plasma between the anode and the first cathode and a resulting ionic current flowing into the first cathode;

a second cathode electrically isolated from the first cathode and spaced from the anode by a second spacing less than the first spacing, the second spacing sufficient to enable formation of a plasma between the anode and the second cathode and a resulting ionic current flowing into the second cathode;

a magnet applying a magnetic field across the first and second spacings to lengthen a free electron path to sustain the plasma; and

an electrical controller applying a voltage between each of the first and second cathodes and the anode to generate plasma discharge ionization between the anode and the first and second cathodes within respective pressure ranges, the electrical controller determining the pressure from the measured current flowing to the first cathode and from the measured current flowing to the second cathode.

41. The cold cathode ionization gauge of claim 40 wherein the electrical controller determines a first cathode impedance and a second cathode impedance from the measured current flowing to the first cathode, the measured current flowing to the second cathode, and the measured anode voltage, and pressure is determined from the first cathode impedance and the second cathode impedance.